Word Senses and Semantic Representations Can We Have Both?

Processing of metaphor has long been studied due to implications on human cognition. But less has been explored as to its relationship with lexical access. This is important, because comprehension of conventional metaphors fundamentally hinges on activation of lexicalized metaphorical meanings, but early lexical access models rarely consider a distinction between literal and metaphorical meanings. This study attempts to bring these two lines of research together by investigating processing of lexical metaphors, i.e., metaphorical polysemy (e.g., “pressure”) based on Giora’s (2003) graded salience hypothesis. The graded salience hypothesis states that salient meanings, i.e., “more frequent, more familiar, or more conventional” meanings, are always initially accessed in context. Giora’s model differs from previous models because it comprehensively interprets lexical access as well as figurative language processing, thus receiving much attention in literature (Gibbs, 2001). But it has two major problems: (1) Vague definition of salience, particularly its relationship with frequency, and (2) Unclear criteria for semantic analysis and meaning distinction. To address these issues, this thesis combined a corpus linguistic approach and recent cognitive semantic theories with psycholinguistic experiments. I distinguished sense-level meanings from facet-level meanings based on Chinese Wordnet and calculate their sense frequencies in corpus. In one offline word association task and two online cross-modal lexical priming tasks, I found: Salient, sense-level literal meanings cannot be sufficiently activated in metaphorically-biased contexts. This contradicted the graded salience hypothesis’s prediction but corresponded to Ahrens et al.’s (1998) and Croft and Cruse’s (2004) lexical semantic theories that sense-level meanings cannot coexist in the same context. The study thus has significant implications on the graded salience hypothesis and for metaphor and lexical access research in general; particularly, a refined lexical semantic analysis based on cognitive linguistic theories is necessary. Besides, the results shed light on the relationship between frequency and salience, i.e., how corpus reflects human cognition, as well as how cognitive semantic theories can refine interpretations in psycholinguistic studies. Therefore, the study is expected to have a cross-domain contribution to the general field of cognitive science.

A variety of applications have been built in recent years with the aim to extract knowledge from Al-Quran. Current knowledge representations of Al-Quran give attention primarily on conceptual ontology models that describe the semantic relations between the Quranic concepts or entities. There seems to be minimal effort towards recognizing the semantic relations between words in Quranic text, which is relatively more complex. This paper aims to present a framework for semantic knowledge representation of Al-Quran using dependency relations between words, in an attempt to boost the retrieval accuracy for Al-Quran. The semantic analysis is performed on Quranic verses according to word dependency relations using dependency parsing. Based on parsed dependencies, a set of rules are formulated to build a semantic graph of Surah Ali Imran of Al-Quran. The efficiency of the semantic representation was tested by developing a prototype question answering system. The framework was evaluated using precision and recall, First Hit Success, First Answer Reciprocal Rank and Total Reciprocal Rank by comparing the retrieved and actual answers. The results indicate that the performance of the proposed framework using word dependencies improves the semantic representation of knowledge.

Goal. Comment on ethno-topo-hydroantracite and other terms and names in the book of A. N. Tarasova, Timur in the Caucasus. Methods. Used the comparative-historical method; methods of linguistic analysis linguistic material and, above all, the analysis of lexical semantics; methods of comparative historical linguistics and structuralism. Results. Commented ethno-topo-hydroantracite and other terms and names in the book of A. N. Tarasova, Timur in the Caucasus. Conclusion. Using comparative-historical methods; methods of linguistic analysis linguistic material and, above all, the analysis of lexical semantics; methods of comparative historical linguistics and structuralism has shown its effectiveness to review the ethno-topo-hydroantracite and other terms and names in the book of A. N. Tarasova, Timur in the Caucasus.

A growing interest in tasks involving language understanding by the NLP community has led to the need for effective semantic parsing and inference. Modern NLP systems use semantic representations that do not quite fulfill the nuanced needs for language understanding: adequately modeling language semantics, enabling general inferences, and being accurately recoverable. This document describes underspecified logical forms (ULF) for Episodic Logic (EL), which is an initial form for a semantic representation that balances these needs. ULFs fully resolve the semantic type structure while leaving issues such as quantifier scope, word sense, and anaphora unresolved; they provide a starting point for further resolution into EL, and enable certain structural inferences without further resolution. This document also presents preliminary results of creating a hand-annotated corpus of ULFs for the purpose of training a precise ULF parser, showing a three-person pairwise interannotator agreement of 0.88 on confident annotations. We hypothesize that a divide-and-conquer approach to semantic parsing starting with derivation of ULFs will lead to semantic analyses that do justice to subtle aspects of linguistic meaning, and will enable construction of more accurate semantic parsers.

In order to overcome the manpower and time problems of two types of syntactically similar program conversion systems, this paper proposes an automatic conversion system for two types of syntactically similar programs based on the compilation principle. The method is divided into three stages: lexical analysis, syntactic analysis and semantic analysis. The lexical analysis scans the source program character by character from left to right and generates a sequence of words for syntactic analysis. Syntactic analysis further groups the symbols of the source program to generate a syntax tree representing the structure of the source program. Semantic analysis traverses the syntax tree to find the rule base for implementing the source program. Finally the syntax tree is traversed to generate the target program. In creating the syntax tree, the composition mode is used, while in traversing the syntax tree, the visitor mode is used. The experimental results show that these two methods can reduce the time of manual conversion.

An incremental basic block instruction scheduler

The development of Question Answering (QA) systems has become important because it reveals research issues that require insight from a variety of disciplines, including Artificial Intelligence, Information Extraction, Natural Language Processing, and Psychology. Our goal here is to develop a QA approach to answer yes/no questions relevant to civil laws in legal bar exams. A bar examination is intended to determine whether a candidate is qualified to practice law in a given jurisdiction. We have found that the development of a QA system for this task provides insight into the challenges of formalizing reasoning about legal text, and about how to exploit advances in computational linguistics. We separate our QA approach into two steps. The first step is to identify legal documents relevant to the exam questions; the second step is to answer the questions by analyzing the relevant documents. In our initial approach described here, the first step has been already solved for us: the appropriate articles for each question have been identified by legal experts. So here, we focus on the second task, which can be considered as a form of Recognizing Textual Entailment (RTE), where input to the system is a question sentence and its corresponding civil law article(s), and the output is a binary answer: whether the question sentence is entailed from the article(s). We propose a hybrid method, which combines simple rules and an unsupervised learning model using deep linguistic features. We first construct a knowledge base for negation and antonym words for the legal domain. We then identify potential premise and conclusion components of input questions and documents, based on text patterns and separating commas. We further classify the questions into easy and difficult ones, and develop a two-phase method for answering yes/no questions. We answer easy questions by negation/antonym detection. For more difficult questions, we adapt an unsupervised machine learning method based on morphological, syntactic, and lexical semantic analysis on identified premises and conclusions. This provides the basis to compare the semantic correlation between a question and a legal article. Our experimental results show reasonable performance, which improves the baseline system, and outperforms an SVM-based supervised machine learning model.

Artificial Intelligence (AI) has enabled machines to perform tasks traditionally carried out by humans, including the creation of written works such as poetry. Exploring the emotional power, aesthetic dimensions, and embedded meanings contained in poetry generated by AI offers a compelling basis for comparison with poetry by human authors. This research aims to assess and compare the critical strength of poetry generated by AI (Gemini, Perplexity, and ChatGPT) with that of literary works, using a sample of social criticism-themed poetry by W.S. Rendra, Wiji Thukul, and Gus Mus. It employs descriptive research with a comparative approach using lexical and referential semantic methods. The findings reveal that the social criticism poetry of W.S. Rendra, Wiji Thukul, and Gus Mus possesses a deeper and sharper critical and emotional strength compared to that created by AI (Gemini, Perplexity, and ChatGPT). The poetry by AI is unable to surpass the emotional power of humans. Although the themes of the poetry requested in the three AI applications are pretty complex and detailed, the results sometimes deviate from expectations. The language and diction used tend to be generic and occasionally irrelevant, which affects the meaning of the poetry, particularly in terms of lexical and referential meanings.

The article discusses the lexical and cultural semantics of religious words. The cultural semantics of religious words is revealed in connection with the knowledge, faith, and national existence of the people and additionally includes information of religious content. The reason is, traditional religious knowledge is the core of our national identity. It is known that the daily life, customs, beliefs and existence of the Kazakh people are connected with the religion of Islam. In this regard, religious words are found in oral and written literature, fairy tales, proverbs and sayings. The article analyzes the religious words found in the works of “Word of the Ancestors”. Descriptive, ethnolinguistic, lexico-semantic methods were used. The lexical and cultural semantic analysis of religious words in the article is of theoretical and practical importance. In particular, it is useful research for creating a Linguistic and Cultural dictionary, replenishing the National Corpus Base, and the work of doctoral and undergraduate students.

Recent research works on unsupervised word sense disambiguation report an increase in performance, which reduces their handicap from the respective supervised approaches for the same task. Among the latest state of the art methods, those that use semantic graphs reported the best results. Such methods create a graph comprising the words to be disambiguated and their corresponding candidate senses. The graph is expanded by adding semantic edges and nodes from a thesaurus. The selection of the most appropriate sense per word occurrence is then made through the use of graph processing algorithms that offer a degree of importance among the graph vertices. In this paper we experimentally investigate the performance of such methods. We additionally evaluate a new method, which is based on a recently introduced algorithm for computing similarity between graph vertices, P-Rank. We evaluate the performance of all alternatives in two benchmark data sets, Senseval 2 and 3, using WordNet. The current study shows the differences in the performance of each method, when applied on the same semantic graph representation, and analyzes the pros and cons of each method for each part of speech separately. Furthermore, it analyzes the levels of inter-agreement in the sense selection level, giving further insight on how these methods could be employed in an unsupervised ensemble for word sense disambiguation.KeywordsSense NodeSemantic NetworkWord Sense DisambiguationSemantic GraphCorrect SenseThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Requirements engineering's continuing dependence on natural language description has made it the focus of several efforts to apply language engineering techniques. The raw textual material that forms an input to early phase requirements engineering and which informs the subsequent formulation of the requirements is inevitably uncontrolled and this makes its processing very hard. Nevertheless, sufficiently robust techniques do exist that can be used to aid the requirements engineer provided that the scope of what can be achieved is understood. In this paper, we show how combinations of lexical and shallow semantic analysis techniques developed from corpus linguistics can help human analysts acquire the deep understanding needed as the first step towards the synthesis of requirements.

Introduction. Educational and methodological interaction between teacher and student during lecture caused many questions in recent years. The essential structural part of this interaction is work with text. Questions arise relative to studentsТ inability to find core information in text, to clarify the concepts, to explain the logics of presentation, to understand syntax of a course-book or lecture text. The research problem is to integrate pedagogical and psycholinguistic approaches and to develop new mechanisms for working with text. The purpose of this article is to develop integrative mechanism s for working with academic texts. Materials and Methods. In the study we drew on analysis and comparison, induction and deduction, psycholinguistic analysis, and generalisation. The methodology of working with text was developed based on structuralism, hermeneutics and the theory of unity. Kazan linguistic school (founded by Prof. I. A. Baudouin de Courtenay) considers a text as a historical and psychological phenomenon: the text materialises a set of methods of cognition. This approach was successfully developed by German psycholinguists. Therefore, the theoretical basis of our research is technique by D. Hake, K. Nestler (Germany) adapted for evaluating the effectiveness of pedagogical communication. Results. The study developed mechanisms for increasing the efficiency of the teaching and methodological interaction between teacher and student based on integration of traditional methods of working with text in university and psycholinguistic procedures of text analysis. Integration of these methods with psycholinguistic procedures of information, logical interrelation, conceptual and syntactic complexity analysis made it possible to improve efficiency of the text work and consequently - educational and methodological interaction. As a result we identified four mechanisms: testing the structural-semantic content, improvement of structural and logical analysis, mastering the procedure of lexical and conceptual analysis, structural-syntactical analysis. The results of the study have practical significance: the mechanisms can be reproduced in a wide range of high-school practice and affect the quality of education, improve the level of social and intercultural communications. Discussion and Conclusions. In the educational practice of universities, integrative mechanisms to increase the efficiency of educational and methodological interaction are not sufficiently applied. Our results suggest that the use of integrative mechanisms allows improving the efficiency of not only text work, but also the quality of teacher training in general. The findings of the study will be useful to students, teachers and specialists in the field of education and pedagogical communica tion.

This paper accounts for the suffix -ang in Balinese and it focuses on its syntactic and semantic representation. Using I Madé’s Sugianto’s Ki Bari Gajah, a one hundred fifty-page Balinese novel and informants as the data, and applying the RRG theory by Van Valin and Randy (1999) other thoughts of the experts of Balinese, it was found out that -ang functioning as a transitivizing suffix can attach to noun, adjective, adverbs and verbs and imply various syntactic structures and semantic representations. Suffix -ang attached to the base in imperative sentences express no meaning. In this case, it is just used to imply that the sentence is in the form of imperative. Like other languages, English for example, one derived verb with -ang may be used transitively or intransitively.

Semantic information is important for precise word sense disambiguation system and the kind of semantic analysis used in sophisticated natural language processing such as machine translation, question answering, etc.There are at least two kinds of semantic information: lexical semantics for words and phrases and structural semantics for phrases and sentences.We have built a Japanese corpus of over three million words with both lexical and structural semantic information.In this paper, we focus on our method of annotating the lexical semantics, that is building a word sense tagged corpus and its properties.

Compiler design plays an important role in ensuring that the translation of the programs written in high-level language into executable code is correct. However, in todays’ safety-critical environments, security gaps, visible and hidden defects in compiler models are liability factors that needed to be addressed in the process of examining that a system meets specifications and requirements of its intended objectives. Nevertheless, defects in this model are errors in coding which bring about incorrect or unexpected results from a compiler design that does not meet the actual requirements. Hence, this paper developed a novel approach by integrating the computational power of Communicating Stream X-Machine (CSXM) to address the problem associated with compiler correctness. In addition, this paper outlines the development of a novel compiler design called X-compiler, emphasizing its important role in software development. CSXM technique was implemented in Visual Studio (2022) to develop the X-compiler. The X-compiler utilize state-of-the-art techniques to translate program source code written in high-level programming language which is the blend of C# and Python programming language into executable code. The development process includes the lexical analysis, syntax parsing, semantic analysis, and code generation ensuring the correctness of the X-compiler. The results of the compiled code were run to produce output in command Windows environment for user interactions. Also, highlights the potential for enhanced language interoperability and the development of efficient compiler that leverage the strengths of multiple programming paradigms. Overall, the study on CSXM contributed to a deeper understanding of concurrent systems, while the design and implementation of the compiler showcased the feasibility of creating a synergistic blend of C# and Python programming languages.